Accepted Manuscript

A comparative study of solid carbon acid catalysts for the esterification of free

fatty acids for biodiesel production. Evidence for the leaching of colloidal car‐

bon

Chinmay A. Deshmane, Marcus W. Wright, Abdessadek Lachgar, Matthew

Rohlfing, Zhening Liu, James Le, Brian E. Hanson

PII: S0960-8524(13)01303-5

DOI: http://dx.doi.org/10.1016/j.biortech.2013.08.073

Reference: BITE 12272

To appear in: Bioresource Technology

Received Date: 28 May 2013

Revised Date: 7 August 2013

Accepted Date: 9 August 2013

Please cite this article as: Deshmane, C.A., Wright, M.W., Lachgar, A., Rohlfing, M., Liu, Z., Le, J., Hanson, B.E.,

A comparative study of solid carbon acid catalysts for the esterification of free fatty acids for biodiesel production.

Evidence for the leaching of colloidal carbon, Bioresource Technology (2013), doi: http://dx.doi.org/10.1016/

j.biortech.2013.08.073

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers

we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and

review of the resulting proof before it is published in its final form. Please note that during the production process

errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

*hanson@vt.edu, phone: 540-231-7206, fax: 540-231-3255; lachgar@wfu.edu, phone: 336-758-4676, fax: 336-758-4656.

A Comparative Study of Solid Carbon Acid Catalysts for the Esterification

of Free Fatty Acids for Biodiesel Production. Evidence for the Leaching of Colloidal

Carbon

Chinmay A. Deshmane+, Marcus W. Wright,

+ Abdessadek Lachgar*,

+ Matthew

Rohlfing,+ Zhening Liu,

+ James Le,

+ and Brian E. Hanson*

,++

Contribution from:

+Department of Chemistry, Wake Forest University, Winston Salem, NC

and

++

Department of Chemistry, Virginia Polytechnic Institute and State University,

Blacksburg, VA 24061-0212

Abstract

The preparation of a variety of sulfonated carbons and their use in the

esterification of oleic acid is reported. All sulfonated materials show some loss in

activity associated with the leaching of active sites. Exhaustive leaching shows that a

finite amount of activity is lost from the carbons in the form of colloids. Fully leached

catalysts show no loss in activity upon recycling. The best catalysts; 1, 3, and 6; show

intitial TOFs of 0.07 s-1, 0.05 s-1 , and 0.14 s-1, respectively. These compare favorably

with literature values. Significantly, the leachate solutions obtained from catalysts 1, 3,

and 6, also show excellent esterification activity. The results of TEM and catalyst

poisoning experiments on the leachate solutions associate the catalytic activity of these

solutions with carbon colloids. This mechanism for leaching active sites from sulfonated

carbons is previously unrecognized.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

2

Key words: Esterification; biodiesel; solid carbon acids

1. Introduction

Porous carbon materials are of interest in many applications because of their high

surface area, physicochemical properties, stability, low density, and their wide

availability. They are used extensively as electrode materials for batteries, fuel cells, and

supercapacitors, as sorbents for separation processes and gas storage, and as supports for

many important catalytic processes. In a seminal paper in the area of solid acid catalysis

Toda et al. showed that partially dehydrated sugars are readily sulfonated and can be used

as catalysts for esterification of fatty acids.(Toda et al., 2005) The publication of this

work has spawned an enormous effort to prepare additional solid carbon acids. To date

the utility of these carbon acids however appears to be limited by their tendency to leach

acid sites.(Mo et al., 2008a) Goodwin et al. showed that sulfonated partially carbonized

sugars slowly leach active sites in the form of sulfonated polycyclic hydrocarbons until a

stable solid catalyst with residual catalytic activity is obtained. Further the partially

carbonized materials have low surface areas and swelling of the catalysts is required for

good catalytic activity. Very recently it has been suggested that deactivation occurs by

sulfate ester formation on the surface of carbon acids. (Fraile et al., 2012)

In response to fossil fuels depletion, soaring fuel prices, concern over greenhouse

gas emissions, and the geopolitics of oil, biodiesel has regained traction as an alternative

fuel that can be used directly in diesel engines, heating systems, and electricity

generators.(Chen & Fang, 2011; Hara, 2010; Lou et al., 2011; Rao et al., 2011; Shu et al.,

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

3

2010; Toda et al., 2005; Tropecelo et al., 2010) Biodiesel is a minor component of US

transportation fuels; in 2007 over 200 billion gallons of transportation fuels were

consumed in the US while total biodiesel production in the US was approximately 450

million gallons the same year. Biodiesel production in the US grew to more than 1 billion

gallons in 2011. European production of biodiesel is significantly higher, exceeding 2.8

billion gallons. (Data from the US Bureau of Transportation Statistics, www.bts.gov,

www.biodiesel.org and the European Biodiesel Board www.ebb-eu.org). Biodiesel,

comprised of esters of fatty acids that are obtained from triglycerides in vegetable and

animal oils, is a drop-in, renewable replacement for diesel. Alternatively, biodiesel can be

used as an additive to petroleum diesel to improve the lubricity of the fuel. (Hu et al.,

2005) In addition to biodiesel’s use as a fuel, esters of fatty acids are precursors to a

variety of commercial chemicals. (Akoh & Swanson, 1989; Donnelly & Bulock, 1988;

Jandacek, 1991 ) Biodiesel is similar to petroleum diesel in combustion properties, but

free of sulfur, making it a cleaner burning fuel than petroleum diesel. (Cantrell et al.,

2005) It is derived from renewable sources and is biodegradable. Further, it has a high

flash point and lower pollutants emissions compared to conventional fossil fuels. (Aricetti

& Tubino, 2012; Zabeti et al., 2010)

To make biodiesel, triglycerides are transformed to mono esters in a base catalyzed

transesterification process that does not function well in the presence of free fatty acids

(FFAs).(Leclercq et al., 2001; Lotero et al., 2005) High quality vegetable oil feedstock,

with low FFA content, is expensive and represents much of current biodiesel production

costs. To be more cost competitive new processes for the formation of biodiesel from

inexpensive feedstock, which contain large amounts of free fatty acids, need to be

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

4

developed. Thus, the development of catalysts for the esterification of acids continues to

be an area of active research. Recently it has been shown that the sulfonation of corn

straw derived carbon can be sulfonated to yield a high density of sulfonic acid groups

(Liu et al., 2013) and that the rate of esterification of butyric acid decreases as the carbon

number of the alcohol used to prepare the ester increases.(Pappu et al., 2013)

The presence of FFAs in the base catalyzed transesterification step leads to the

formation of soaps, which are difficult to separate and cause a decrease in biodiesel yield.

It is generally accepted that a two-step process is required to treat oils with a high FFA

content; in the first step an acid-catalyzed esterification of FFAs is accomplished and in

the second step the base catalyzed transesterification reaction is carried out. Both

reactions are done homogeneously under relatively mild conditions. In principle a single

acid catalyzed reaction is possible however the acid catalyzed transesterification reaction

is kinetically slow and requires high temperatures. The homogeneously catalyzed

reactions required for the treatment of oils with high FFA content generate significant

quantities of corrosive aqueous waste. (Srivastava & Prasad, 2000) Thus there is a real

need for effective solid catalysts for the esterification and transesterification reactions.

Ideally a heterogeneous catalyst would perform both reactions under mild conditions, but

in the absence of achieving that goal, separate solid catalysts for the two-step process are

desirable.

A comparison of several carbon acids and evaluate their relative stability toward

leaching of acid sites is reported here. The carbon catalysts prepared in this work include

low and high surface area carbons derived from sugars, polymers, and high surface area

silica-templated carbons. Recent work by Goodwin et al. (Lopez et al., 2005; Mo et al.,

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

5

2008a; Mo et al., 2008b) and Fraile et al. (Fraile et al., 2012) show the importance of

understanding the nature of the solid acid leaching and deactivation. Goodwin ascribes

the deactivation to the leaching of sulfonated polycyclic hydrocarbons and Fraile et al.

suggest that the solid acid itself is deactivated by the formation of sulfonate esters on the

surface. NMR data on leached solutions supports the idea that sulfonated polycyclic

hydrocarbons are present in the leachate (Mo et al., 2008a) and magic angle spinning

NMR data on deactivated solids are consistent with the presence of chemisorbed

alcohols.(Fraile et al., 2012) Similar to Goodwin et al. the recycled solid catalysts here

come to a stable level of activity that is consistent with permanently stable acid sites on

the solid. However, the nature of the leachate is most consistent with sub micron colloids

in solution and these colloidal suspensions have an excellent activity for the esterification

of free fatty acids.

2. Experimental Section

2.1 General: Distilled deionized water was used for all synthetic operations in water.

Reagent grade methanol was dried by passing the solvent through aluminium oxide

drying columns immediately prior to use. Reagents, glucose, ZnCl2.H2O and

concentrated sulfuric acid were obtained from Fisher Scientific Company. Elemental

analyses were obtained from Atlantic Microlabs. Powder X-Ray diffraction data were

collected on a Bruker D2 desktop diffractometer with CuKÅ. Surface area

measurements were recorded using an Autosorb iQ gas sorption analyzer. SEM and

HRTEM images of the samples and carbon colloids were obtained with JEOL (JSM-

6330F) field emission scanning electron microscope and JEOL (JEM-1200EX) electron

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

6

microscope respectively. Thermogravimetric analysis (TGA) and infrared absorption

spectra were recorded on Perkin Elmer Pyris 1 TGA and Perkin Elmer FTIR Spectrum

100, respectively. TGA analyses were carried out on 10 mg samples that were heated in

an inert atmosphere of Argon at heating rate of 40˚C/min up to 700˚ or 900˚C.

GC-MS spectra were collected on an Agilent 7890A GC with a 5975C MSD

running an Agilent 30m x 0.250 mm 0.25 m HP-5MS column. Samples were typically

prepared by diluting 10 L of the reaction mixture into 1 mL HPLC grade hexane.

Injections via the auto sampler were 1 L, split 50:1 at a constant 1.2 mL/min helium

flow. The injector was held at 250 C with the oven holding at 50 C for 2 minutes then

ramping to 320˚C at 25 degrees per min. The total ion chromatograms were integrated

and relative integrations found.

2.2 Preparation of Sulfonated Carbons:

Catalyst 1. Hydrothermal Carbons (HTC): These are prepared from aqueous

glucose solutions (1 to 3 molar) at 200°C in teflon lined acid digestion autoclaves. In a

typical synthesis procedure, 10 g of glucose was dissolved in 20 ml of distilled water and

transferred to a Teflon line autoclave and placed in an oven at 200˚C for 24h. The

carbonaceous material after the HTC step is separated by filtration, washed with distilled

water (3 x 30 ml) followed by an ethanol wash (1 x 30 ml) and placed for drying

overnight at 100˚C. A yield of 4 g was obtained.

The as-prepared “carbons” contain a high percentage of hydrogen and oxygen;

total carbon content is typically in the range of 60 to 70 weight percent. The surface area

of HTCs are low, generally less than 10 m2g

-1. Sulfonation of HTCs (S-HTC) leads to

materials that contain 53% carbon and up to 1.45 % sulfur. Significantly, sulfonation of

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

7

HTCs increases the surface area of these materials to a range of 300 to 600 m2g

-1.

Analytical data for all carbons is collected in Table 1.

Catalyst 2. Pyrolyzed HTCs (p-HTC): The as-prepared HTC carbons are

pyrolyzed under argon at 850°C. These materials are 95 % carbon and contain very little

residual hydrogen, 0.3 %. Pyrolysis leads to an increase in surface area to ca 300 m2g

-1.

Sulfonation of p-HTC (S-p-HTC) leads to materials that contain 2.25 % sulfur. Surface

areas of the pyrolyzed carbons typically decrease upon sulfonation.

Catalyst 3. Hydrothermal Carbonization of glucose and polyacrylic acid (HTC-

PAA): Carbon rich in carboxylic acid groups were prepared by the hydrothermal

carbonization of glucose in the presence of acrylic acid. In order to obtain materials with

high degree functionality 2.5 wt% (with respect to glucose) of polyacrylic acid was added

to the reaction mixture. In a typical reaction 10 g glucose and 250 mg acrylic acid were

disolved in 20 mL distilled water and transferred to a Teflon lined autoclave for

hydrothermal treatment as described for catalyst 1. Similarly yield of 4 g was obtained

after the hydrothermal carbonization step. Hydrothermal carbonization of glucose in the

presence of acrylic acid leads to the formation of spherical micron sized spheres that

contain 67% C. Upon sulfonation the carbon content drops to 54 % and a sulfur content

of 1.93 % was determined by elemental analysis of the resulting material.

Catalyst 4. Pyrolyzed HTC carbon prepared with ZnCl2 and CO2 activation: This

carbon material was prepared by employing chemical activation methods in order to

obtain large surface area carbons. A 20 mL glucose/ZnCl2 solution (2.8 and 0.83 M

respectively) was subjected to hydrothermal carbonization at 200˚C. The ZnCl2 to

glucose molar ratio was 0.3. A yield of 3 g of the zinc activated carbon was obtained.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

8

This material was then pyrolyzed at 800˚C under flowing CO2. The HTC carbon was

heated to 800C under argon and then the flow was switched to CO2. Heating was

continued for 5h under CO2 then the system was cooled under argon. These carbons

typically give surface areas in the range of 500 m2g

-1. The pyrolyzed carbon was

sulfonated at 150˚C for 15h, after which the surface area dropped to ca 1 m2/g however

exceptional sulfur content was achieved.

Catalyst 5. High Surface Area Carbons: Pyrolized HTC carbons displaying high

surface area ~ 1300 m2g

-1 were prepared by post synthesis activation with ZnCl2 and CO2.

To achieve this, a zinc chloride solution (100 mg in 0.25 mL water) was added to a 1 g

sample of HTC carbon prepared according to catalyst 1. The sample was then pyrolyzed

at 800°C under CO2 for approximately 5 h. Sulfonation was carried out as described

below.

Catalyst 6. Pyrolized organosiloxane resin: Pyrolysis of (PhSiO1.5)n followed by

base treatment leads to high surface area carbons. The (PhSiO1.5)n resin was prepared by

the acid hydrolysis of PhSi(OMe)3. In a typical preparation, 47.5 mL PhSi(OMe)3 in 100

mL methanol was combined with 122 mL 2.67 M HCl(methanol). After 24 h the

methanol is decanted to yield 30 g resin which was washed methanol and dried in air.

The resulting resin was pyrolyzed at 900˚C under a slow nitrogen purge to yield a carbon

silica composite with a yield of 85%. This material was then pulverized in a mortar to

generate a powder. Extraction of 1 gm of the carbon silica composite in 2 ml 18.8 M

NaOH at 120˚C is carried out in Teflon lined autoclave for 8h. The material was washed

with water until the pH of the washings was 7 and dried in air at 10°C, yield 0.48 g.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

9

Surface areas of 800 to 1200 m2g

-1 were typically achieved. Sulfonation with hot sulfuric

acid, see below, leads to a material that contains 1.53 wt% sulfur.

2.3 Sulfonation procedure: Sulfonation of all carbons prepared above was accomplished

by heating 1 g of carbon in 10 mL concentrated sulfuric acid in a round bottom flask at

150°C for 15 hours. After sulfonation the reaction was quenched by pouring the mixture

into cold water. The solid was collected by vacuum filtration and washed with hot

distilled water until the filtrate no longer precipitates barium sulfate when treated with

Ba(NO3)2.

2.4 Extraction with methanol: Sulfonated carbons are well known to lose activity upon

recycling of the catalyst or by extracting the catalysts with solvents. In an effort to obtain

solids with stable catalytic activity, the carbons were subjected to exhaustive extraction in

a Soxhlet extractor. In all cases 1.0 g of sulfonated carbon was treated with 20 mL

methanol in the Soxhlet extractor under reflux for 4 hours. The catalyst was recovered

and dried. The catalytic activity of both the leachate and the solid catalyst was evaluated

in the esterification of oleic acid. The remaining catalyst was subjected to up to three

more rounds of exhaustive leaching.

2.5 Catalytic activity studies. Catalytic activity studies were performed using the

original sulfonated carbons, the carbons obtained after removal of all leachable sites, as

well as the leachates. This process allows for determining and comparing intrinsic

catalytic activity of the solid acids. The esterification of higher fatty acids was carried out

under the reflux conditions in a methanol-fatty acid mixture with MeOH to FFA molar

ratio of 10, and 10 wt% of catalysts (relative to fatty acid) were used in the reactions. All

catalysts were dried at 100 °C for 2 h prior to the reaction in order to remove adsorbed

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

10

water. The activity of the leachate solutions were determined by taking 1.4 mL of the

methanol solution obtained after leaching for the esterification of 1 g oleic acid. The

esterification products were analyzed by a gas chromatograph mass spectrometer with

capillary column. The colloids have a comparable activity to that of 0.01 vol % sulfuric

acid (1.9 x 10-3

M H2SO4 in methanol) as estimated by monitoring conversion with

respect to time.

Mercury poisoning of the catalysts was accomplished simply by adding a drop of

mercury to the reaction mixture after the addition of all reagents and prior to bringing the

system to reaction temperature.

3. Results and Discussion

The sulfonation of carbon materials is accomplished with concentrated sulfuric

acid and is assumed to lead the formation of C-S bonds at sp2 carbons that are in graphitic

domains in the material. (Fukuhara et al., 2011) It is also possible that sulfate ester

groups containing C-O-S bonds are formed. While this is an oversimplification of the

sulfonation process it is useful to think of the surface acidity that results from the reaction

of carbon with H2SO4 as yielding C-SO3H and C-O-SO3H groups. A simple pH titration

of the sulfonated carbons suspended in aqueous sodium chloride is consistent with the

presence of strong acid functionality although the presence of small quantities of

carboxylic acid or phenolic groups cannot be ruled out.

Data shown in Table 1 indicate that the as-prepared hydrothermal carbons

contain relatively large amount of hydrogen and oxygen corresponding to residual

hydroxyl (OH) and carboxylate (COOH) functionality. For example, the precursor to

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

11

catalyst 1, a hydrothermal carbon obtained from glucose, has an elemental composition of

C5.5H4.5O1.8. These values are more in line with hydrocarbons rather than elemental

carbon. Indeed it has been shown by solid state NMR that hydrothermally prepared

carbons have significant furan content.(Demir-Cakan et al., 2009) The empirical formula

of the catalyst 1 obtained after sulfonation was found to be C4.4H3.0O2.65S0.04. Again, the

material has an elemental compositon similar to hydrocarbon materials. Significantly,

the increase in oxygen content is much greater than can be explained by sulfonation.

Other functional groups, e.g. alcohols and carboxylic acids, must be incorporated by

oxidation upon reaction with sulfuric acid. This increase in oxygen content is also

observed after sulfonation of pyrolyzed materials which have carbon contents that exceed

95%. For example, the sulfonation of the pyrolized carbon precursor to catalyst 2 goes

from a composition of C8.1H0.3O0.16 to C5.3H0.6O0.78S0.07. Again, the increase in oxygen

content exceeds that expected from simple sulfonation. Further, a consequence of

sulfonation is a decrease in the percent carbon in catalysts.

The largest surface areas are observed in catalysts 5 and 6 and these are preserved

upon sulfonation. In catalyst 5 zinc chloride treatment during carbonization leads to

formation of a stable high surface area carbon and in 6 the chemical removal of the in situ

SiO2 template leads to the stable high surface area. The modestly high surface areas

obtained by carbonizing HTC catalysts, materials 2 and 4, collapse upon sulfonation.

Sulfonation of an as-prepared HTC material derived from glucose (catalyst 1) leads to an

increase in surface area whereas an sulfonation of an HTC material from glucose and

PAA leads to a decrease in surface area. The carbon-like network obtained with PAA

incorporation may be less rigid than the material derived from glucose alone.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

12

The TGAs of the prescursors to catalysts 1, 2, 3, and 6, are shown in the

supplemental data. The fully carbonized materials, 2 and 6, show good thermal stability

up to 600 °C. Rapid heating in the TGA under nitrogen for the HTC materials 1 and 3

leads to a loss of just over 90% of the mass by 600 °C. This suggests that rapid pyrolysis

eliminates small molecule hydrocarbons whereas bulk pyrolysis (e.g. to prepare catalyst

2), which is done at a much slower heating rate, leads to carbonization rather than the

elimination of volatile hydrocarbons.

Infrared spectroscopy of the sulfonated HTC carbons shows the presence of

sulfonate groups at 1030 and 1150 cm-1

. These absorption bands are also present in the

IR spectra of the solids obtained after extensive leaching in methanol indicating the

presence of residual sulfonate functional groups. Representative infrared spectra of the

catalysts are shown in the supplementary materials. Powder X-Ray diffraction shows the

presence of amorphous carbon in the HTC carbons. The as prepared hydrothermal

carbons materials show a broad peak at 2=22° which shifts to a value of 24° upon

sulfonation, a value that is more consistent with amorphous carbon. The carbon obtained

from the pyrolyzed siloxane resin shows the presence of both amorphous and graphitic

carbon by powder diffraction with broad peaks at 24 and 44°. Representative powder

diffraction patterns are given in the supplementary material.

3.1 Esterification reactions: The carbon catalysts were screened for activity and

stability toward leaching in batch reactions by comparing conversions at a fixed time.

The esterification of fatty acids is an equilibrium process thus the batch reactions at high

conversion are not indicative of relative rate. However, since the reactions do not reach

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

13

equilibrium, conversion gives a rough estimate of activity. Batch reactions for a fixed

period of time were used to screen the catalysts. Recycling of the catalysts and the

leaching experiments give an indication of the stability of the supported acid catalysts.

The materials which showed the best behavior in terms of stability toward leaching in the

batch reactions, catalysts 1, 3, and 6, were chosen for further study by comparing initial

rates and investigation of the respective leachates (vide infra). The results for the batch

reactions on catalysts 1, 3, and 6, are shown in Figure 1. The results for catalysts 2, 4, and

5, are shown in Figures S1, S2, and S3 in the supplementary materials.

The results from the esterification of oleic acid with catalyst 1 is shown in Figure

1A. Run 0 shows the result of the freshly sulfonated catalyst while Runs 1 through 3 are

after extracting the catalyst one, two, and three times, respectively, with hot methanol as

described in the experimental section. The cross-hatched activity bars are the results

from the methanol leachate.

Catalyst 2 was prepared from an HTC carbon that was pyrolyzed before

sulfonation. Pyrolysis lead to an increase in surface area (see Table 1) but surprisingly

sulfonation lead to a collapse in surface area to less than 1 m2g

-1. It spite of the low

surface activity the catalyst still had respectable activity (see supplementary material,

Figure S1).

Catalyst 3 was prepared by the hydrothermal carbonization of glucose in the

presence of polyacrylic acid. The addition of PAA to glucose has been described in the

literature to generate partially carbonized materials at 400 °C that have increased acid

functionality.(Demir-Cakan et al., 2009) Compared to the as-prepared HTC carbon from

glucose, catalyst 1, the addition of PAA doesn’t increase oxygen content. The catalytic

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

14

results obtained with 3 are shown in Figure 1B. After two extractions with methanol the

leachate activity is in line with the activity without catalyst while the solid continues to

show excellent activity.

The addition of zinc chloride prior to the pyrolysis of carbons is known to

increase the surface area of activated carbons.(Ahmadpour & Do, 1997) Catalyst 4 was

prepared in a similar fashion as catalyst 3 except for the addition of zinc chloride at a

Zn:glucose mole ratio of 0.3 to the mixture in the hydrothermal step. Catalyst 4 obtained

upon sulfonation has surface area that is similar to 3, however, the catalytic behavior of 4

is different. The leachate is significantly more active for the esterification reaction even

after the third extraction (see Figure S2 in the supplementary material).

The method used to prepare catalyst 4 in the presence of zinc chloride in the

hydrothermal step failed to yield a material with a significantly higher surface area than

in the absence of zinc and failed to give a solid catalyst that did not leach activity. As an

alternative, a catalyst was prepared in which zinc chloride was added by incipient

wetness to an already prepared HTC carbon. Thus catalyst 5 was prepared from a carbon

similar to catalyst 1 followed by the incorporation of ZnCl2, at 10 wt % ZnCl2, by

incipient wetness followed by pyrolysis at 850°C. This yields a carbon with a surface

area of 1300 m2g

-1, and retains a high surface area of 1160 mg

-1 after sulfonation. As seen

from the esterification results with 5 (Figure S3 in the supplementary material) the solid

catalyst loses most of its catalytic activity after two extraction cycles with methanol

indicating that surface acid sites are easily leached or poisoned on this catalyst.

Hard templating of high surface area mesoporous carbons is often accomplished

by hydrothermal treatment of sugars in presence of mesoporous silicas that are, in turn,

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

15

templated by surfactants.(Titirici et al., 2007) Removal of the silica after high

temperature carbonization by treatment with HF or aqueous base leads to high surface

area carbons. A simplified procedure based on the carbonization of siloxane oligomers

leads to carbons with surface areas of 800 to 1200 m2g

-1.(Hanson et al., 2009) Catalyst 6,

with surface area of 960 m2g

-1 after removal of residual silica, was prepared from

pyrolyzed phenylsiloxane oligomer. Sulfonation of this carbon yielded a material that

retains its high surface area, 880 m2g

-1, and has 1.53 wt% S. The catalytic results with 6

are shown in Figure 1C. Similar to catalyst 3 this catalyst retains good activity after

exhaustive leaching with methanol. After three extractions the leachate shows only

background activity while the solid retains over 90% of its original activity.

3.2 Catalyst recycling and mass transfer limitations of the leached catlaysts. In a

typical catalyst recycling experiment a catalyst that had been extracted (leached) three

times was used in an esterification reaction as described above then the catalyst was

collected, dried and used again for another esterification. Two cycles after three

extractions were done on catalysts 1, 3, and 6. The results were as follows: For catalyst

1, the initial conversion after 3x extraction was 77% in eight hours, the second and third

recyles gave 78 and 80% conversion respectively. For catalyst 3, the conversions were

80, 83, and 84% respectively, and for catalyst 6, the values were 85, 88, and 86%

respectively. These results indicate that not only robust catalysts can be obtained after

proper leaching but that the catalysts maintain their activity after multiple use. Catalysts2,

4, and 5, show either continued leaching or very little activity after several leaching

cycles and therefore were not pursued further.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

16

Arrhenius plots for catalysts 1, 3, and 6, yield activation energies of 11±2, 8±1,

and 11.0±0.4 kcal/mol respectively. The plot for catalyst 6 is linear and indicative of no

mass transfer limitations. Linearity, however is not a proof that the system is not mass

transfer limited, given the relatively low boiling point of methanol there is only a small

window in which reliable initial rate data can be obtained. The plots for catalysts 1, and

3 show some curvature at higher temperature and mass transfer effects cannot be ruled

out for these catalysts. Thus the ability to completely rule out mass transfer effects for

these catalysts is limited.

3.3 Scanning electron microscopy analysis The carbons formed by hydrothermal

treatment of sugars all have spherical morphology with average diameters of 500 nm to

10 microns depending on reaction conditions and additives as determined by SEM. The

smallest particles, average size of ca 0.5 , are obtained from glucose without any

additives. The basic morphology changes very little upon pyrolysis and/or sulfonation.

Images of particles from catalysts 1 and 3 before and after sulfonation are shown in

supplementary materials; the average particle size of catalyst 3 is about 1.2 microns after

sulfonation. The catalysts from the pyrolysis of siloxane resin have completely different

morphology consisting of randomly shaped and sized particles. The particles have a

rougher appearance after sulfonation. Pores generated by extraction of SiO2 are not

observable by SEM.

3.4 Nature of the leachate. Goodwin et al. showed that the methanol leachate of

sulfonated partially carbonized sugars (e.g. glucose heated to 400°C under nitrogen)

contains aromatic protons by 1H NMR spectroscopy. This is interpreted as the presence

of sulfonated polycyclic hydrocarbons in solution and accounts for the loss of catalytic

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

17

activity of carbon acids as well as activity in the leachate from the catalysts. No proton

signals were observed in the aromatic region of the 1H NMR of the leachate from carbon

catalysts. This negative result does not unequivocally prove the absence of sulfonated

aromatics in solution; rather, it is entirely possible that the concentration of sulfonated

aromatics in the leachate are below the NMR detection limit. Nonetheless, as

demonstrated above, the leached solutions are very active toward the esterification

reaction. In addition to sulfonated aromatics it is possible that sulfate esters, e.g. R-

OSO2OH, are leached. The failure to detect the presence of sulfonated aromatics by

NMR prompted us to consider the hypothesis that colloidal carbon particles might be

suspended in the leachate. To test this hypothesis small drops of methanol leachate

solutions were evaporated directly on the copper grid of a TEM and high magnification

TEM images were taken. These were compared with neat methanol used as blank

solution which was handled identically but without contact with sulfonated carbons.

Representative images are shown in Supplementary Materials. Clearly the leached

solutions contain particles that are absent in the blank. The size of the particles observed

in the TEM is nearly one micron, a size that approaches the as-synthesized carbons.

However the morphology is completely different. The particles in the leachate are

irregular in shape and appear to be aggregates of smaller particles. There is no

discernable structure from the images obtained, to date, on these particles. The leachate

from all carbon catalysts studied in this work show the presence of carbon particles by

TEM.

The formation of highly dispersed carbons in water has previously been obtained

by the direct sulfonation of carbon nanotubes.(Peng et al., 2005) These apparently show

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

18

excellent activity for the esterificaton of acetic acid however initial rates are not reported.

The colloidal materials obtained here have morphologies that are quite different from

carbon nanotubes.

Dynamic light scattering of the leached solutions failed to show the presence of

colloids. Any submicron materials present are in too low a concentration to give

observable scattering in the instrument used.

Depending on the catalyst the methanol leachate solutions and solutions can vary

from colorless to yellow. Solutions from catalyst 6 are routinely colorless and those from

3 have an extremely pale yellow color. The first leachate solutions from most other

catalysts are darker in color.

3.5 Poisoning of the colloid. The catalytic activity of heterogeneous catalysts including

metals on carbon and metallic colloids is readily poisoned by mercury whereas

homogeneous coordination catalysts are typically unaffected by metallic mercury for

many reactions. This is a classic test to demonstrate homogenous catalytic activity for

transition metal catalysts.(Anton & Crabtree, 1983; Jaska & Manners, 2004) There is no

precedent for using the mercury test on a catalysts that do not contain transition metals,

however heterogeneous carbon supported metal catalysts are poisoned by the presence of

mercury. Carbons do have an affinity for mercury and high surface area carbons are used

to eliminate metals from flue gasses in some applications. Further, mercury has a high

surface energy and may attract small particles to the surface of mercury droplets to lower

the surface energy.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

19

The activity of the best solid catalysts above, 1, 3, and 6, their respective leachates,

as well as that of sulfuric acid and para toluene sulfonic acid used as controls were

compared in the presence and absence of metallic mercury. The solids were pre-contacted

with mercury and mercury was added to the leachate and homogenous catalyst solutions.

The percent conversions of oleic acid to methyl oleate after five hours reaction time are

shown in Table 2. In the presence of mercury the solid catalysts show modestly reduced

activities, while the leachate activities drop to background levels. On the other hand the

activity of the homogeneous catalysts is not affected. The homogeneous reaction at 0.01

volume percent sulfuric acid has a comparable concentration to that estimated for the

leachates (see Table 3 below). At this concentration mercury does not poison the

homogeneous sulfuric acid catalyst. These results, in combination with the TEM results,

are consistent with the presence of carbon colloids in methanol extracted from the solid

carbon acids (i.e. leachate) and demonstrate that these the colloidal particles are

reponsible for the esterification reaction.

The persistence of the activity of the solid catalysts suggests that mercury cannot

block access to the active sites in these materials. The mode of action of the mercury on

the colloidal catalysts is unknown at this time. A process analogous to the poisoning of

metallic catalysts cannot be operative since carbons do not react with mercury to form

amalgams. If colloidal particles of carbon aggregate on the surface of the mercury

droplet this would reduce the available active sites for the catalytic reaction. Larger

particles, present when using the bulk solid catalysts, may not have the same affinity for

the mercury droplets and thus the activity of the solids are not impacted to the same

extent.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

20

3.5 Origin of the colloids. The colloidal particles could be generated at any step of the

synthesis procedure. In the case of the HTC carbons, 1-5, the hydrothermal step, the

sulfonation step, or the leaching step could lead to small particles that are suspended in

solution. In the case of 6 colloidal sized particles could be generated at the pulverization

step, the sulfonation step, or the leaching step. For each of the catalysts there appears to

be a finite amount of leaching that takes place and for this reason the leaching step is

unlikely to be the origin of the colloidal particles, rather this is the method of harvesting

the particles. The catalysts are washed with ethanol after hydrothermal carbonization and

again after sulfonation. However, these washing steps are not exhaustive and apparently

do not remove all of the nano scale carbon. Now that the colloids are recognized as part

of the solid carbon acid system it is worth investigating their activity.

3.6 Colloids vs solid catalyst activities. To get an estimate of the activity of the colloids

a good estimate of the sulfur content of the leachates is needed. Sulfur analysis of the

solid catalysts before and after exhaustive (3x) leaching with methanol as described

above showed up to 30% loss of sulfur content. Unfortunately, this does not allow an

accurate estimate of the leachate sulfur content, since the carbon mass balance during the

leaching process is not known. That is, since clearly carbon as well as sulfur is lost in the

leaching process, the leachate can have a significant sulfur content even if the recovered

solid carbon retains the same weight percent sulfur as the beginning catalyst material.

Conversions and initial turnover frequencies obtained are shown in Table 3 for the

solid catalysts after leaching. For comparison, the activity for 0.01 vol% sulfuric acid is

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

21

shown. The values obtained for the solid carbon catalysts are consistent with values

reported in the literature.(Fraile et al., 2012; Peng et al., 2010; Zong et al., 2007) The

superior performance of catalysts 1, 3, and 6, is most likely due to a greater density of C-

SO3H groups and the presence of fewer small particles that can lead to colloidal carbon.

4. Conclusions

The loss of activity upon leaching among the catalysts reported here is due to the

leaching of sub micron sized carbon. The superior behavior of catalysts 1, 3, and 6 is

attributed to more stable SO3H groups and fewer small particles to be leached. The

colloidal material that is leached from the carbons shows activity similar to sulfuric acid

except that the colloids are poisoned by metallic mercury. The carbon colloids have an

affinity for mercury that removes them from the reaction medium and renders them

inactive for catalysis.

Acknowledgements

Support for this work was provided by the Biofuels Center of North Carolina

(Grant No. 2011-104). The powder diffractometer was purchased with funds from NSF

(MRI 201 1040264) awarded to Wake Forest University). BEH acknowledges sabbatical

support from Virginia Tech and the hospitality of the Chemistry Department at Wake

Forest University during his sabbatical visit.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

22

Supplementary Information. Batch catalytic results for catalysts 2, 4, and 5, FTIR

spectra and TGAs of the as synthesized catalysts, sulfonated catalysts and leached

catalyst 1 and 6, and SEM images of catalysts 4 and 5 are available as supplementary

information (6 pages). Supplementary data associated with this article can be found

online.

References

Ahmadpour, A., Do, D.D. 1997. The preparation of activated carbon from macadamia nutshell by chemical activation. Carbon, 35(12), 1723-1732.

Akoh, C.C., Swanson, B.G. 1989. SYNTHESIS AND PROPERTIES OF ALKYL GLYCOSIDE AND STACHYOSE FATTY-ACID POLYESTERS. Journal of the American Oil Chemists Society, 66(9), 1295-1301.

Anton, D.R., Crabtree, R.H. 1983. DIBENZO A,E CYCLOOCTATETRAENE IN A PROPOSED TEST FOR HETEROGENEITY IN CATALYSTS FORMED FROM SOLUBLE PLATINUM GROUP METAL-COMPLEXES. Organometallics, 2(7), 855-859.

Aricetti, J.A., Tubino, M. 2012. A green and simple visual method for the determination of the acid-number of biodiesel. Fuel, 95(1), 659-661.

Cantrell, D.G., Gillie, L.J., Lee, A.F., Wilson, K. 2005. Structure-reactivity correlations in MgAl hydrotalcite catalysts for biodiesel synthesis. Applied Catalysis a-General, 287(2), 183-190.

Chen, G., Fang, B. 2011. Preparation of solid acid catalyst from glucose-starch mixture for biodiesel production. Bioresource Technology, 102(3), 2635-2640.

Demir-Cakan, R., Baccile, N., Antonietti, M., Titirici, M.-M. 2009. Carboxylate-Rich Carbonaceous Materials via One-Step Hydrothermal Carbonization of Glucose in the Presence of Acrylic Acid. Chemistry of Materials, 21(3), 484-490.

Donnelly, M.J., Bulock, J.D. 1988. SUCROSE ESTERS AS MODEL COMPOUNDS FOR SYNTHETIC GLYCOLIPID SURFACTANTS. Journal of the American Oil Chemists Society, 65(2), 284-287.

Fraile, J.M., Garcia-Bordeje, E., Roldan, L. 2012. Deactivation of sulfonated hydrothermal carbons in the presence of alcohols: Evidences for sulfonic esters formation. Journal of Catalysis, 289, 73-79.

Fukuhara, K., Nakajima, K., Kitano, M., Kato, H., Hayashi, S., Hara, M. 2011. Structure and Catalysis of Cellulose-Derived Amorphous Carbon Bearing SO(3)H Groups. Chemsuschem, 4(6), 778-784.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

23

Hanson, B.E., Agaskar, P.A., Hull, R.M. 2009. Patent Application, Single Source Materials to High Surface Area Carbons for Electric Double Layer Capacitors, Pseudocapacitors, and Gas Adsorption.

Hara, M. 2010. Biodiesel Production by Amorphous Carbon Bearing SO(3)H, COOH and Phenolic OH Groups, a Solid Bronsted Acid Catalyst. Topics in Catalysis, 53(11-12), 805-810.

Hu, J.B., Du, Z.X., Li, C.X., Min, E. 2005. Study on the lubrication properties of biodiesel as fuel lubricity enhancers. Fuel, 84(12-13), 1601-1606.

Jandacek, R.J. 1991. DEVELOPING A FAT SUBSTITUTE. Chemtech, 21(7), 398-402. Jaska, C.A., Manners, I. 2004. Heterogeneous or homogeneous catalysis? Mechanistic

studies of the rhodium-catalyzed dehydrocoupling of amine-borane and phosphine-borane adducts. Journal of the American Chemical Society, 126(31), 9776-9785.

Leclercq, E., Finiels, A., Moreau, C. 2001. Transesterification of rapeseed oil in the presence of basic zeolites and related solid catalysts. Journal of the American Oil Chemists Society, 78(11), 1161-1165.

Liu, T., Li, Z., Li, W., Shi, C., Wang, Y. 2013. Preparation and characterization of biomass carbon-based solid acid catalyst for the esterification of oleic acid with methanol. Bioresource Technology, 133, 618-621.

Lopez, D.E., Goodwin, J.G., Bruce, D.A., Lotero, E. 2005. Transesterification of triacetin with methanol on solid acid and base catalysts. Applied Catalysis a-General, 295(2), 97-105.

Lotero, E., Liu, Y.J., Lopez, D.E., Suwannakarn, K., Bruce, D.A., Goodwin, J.G. 2005. Synthesis of biodiesel via acid catalysis. Industrial & Engineering Chemistry Research, 44(14), 5353-5363.

Lou, W., Cai, J., Duan, Z., Zong, M. 2011. Preparation of Cellulose-Derived Solid Acid Catalyst and Its Use for Production of Biodiesel from Waste Oils with High Acid Value. Chinese Journal of Catalysis, 32(11), 1755-1761.

Mo, X., Lopez, D.E., Suwannakarn, K., Liu, Y., Lotero, E., Goodwin, J.G., Lu, C.Q. 2008a. Activation and deactivation characteristics of sulfonated carbon catalysts. Journal of Catalysis, 254(2), 332-338.

Mo, X.H., Lotero, E., Lu, C.Q., Liu, Y.J., Goodwin, J.G. 2008b. A novel sulfonated carbon composite solid acid catalyst for biodiesel synthesis. Catalysis Letters, 123(1-2), 1-6.

Pappu, V.K.S., Kanyi, V., Santhanakrishnan, A., Lira, C.T., Miller, D.J. 2013. Butyric acid esterification kinetics over Amberlyst solid acid catalysts: The effect of alcohol carbon chain length. Bioresource Technology, 130, 793-797.

Peng, F., Zhang, L., Wang, H.J., Lv, P., Yu, H. 2005. Sulfonated carbon nanotubes as a strong protonic acid catalyst. Carbon, 43(11), 2405-2408.

Peng, L., Philippaerts, A., Ke, X., Van Noyen, J., De Clippel, F., Van Tendeloo, G., Jacobs, P.A., Sels, B.F. 2010. Preparation of sulfonated ordered mesoporous carbon and its use for the esterification of fatty acids. Catalysis Today, 150(1-2), 140-146.

Rao, B., Mouli, K.C., Rambabu, N., Dalai, A.K., Prasad, R.B.N. 2011. Carbon-based solid acid catalyst from de-oiled canola meal for biodiesel production. Catalysis Communications, 14(1), 20-26.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

24

Shu, Q., Nawaz, Z., Gao, J.X., Liao, Y.H., Zhang, Q., Wang, D.Z., Wang, J.F. 2010. Synthesis of biodiesel from a model waste oil feedstock using a carbon-based solid acid catalyst: Reaction and separation. Bioresource Technology, 101(14), 5374-5384.

Srivastava, A., Prasad, R. 2000. Triglycerides-based diesel fuels. Renewable & Sustainable Energy Reviews, 4(2), 111-133.

Titirici, M.M., Thomas, A., Antonietti, M. 2007. Replication and coating of silica templates by hydrothermal carbonization. Advanced Functional Materials, 17(6), 1010-1018.

Toda, M., Takagaki, A., Okamura, M., Kondo, J.N., Hayashi, S., Domen, K., Hara, M. 2005. Green chemistry - Biodiesel made with sugar catalyst. Nature, 438(7065), 178-178.

Tropecelo, A.I., Casimiro, M.H., Fonseca, I.M., Ramos, A.M., Vital, J., Castanheiro, J.E. 2010. Esterification of free fatty acids to biodiesel over heteropolyacids immobilized on mesoporous silica. Applied Catalysis a-General, 390(1-2), 183-189.

Zabeti, M., Daud, W., Aroua, M.K. 2010. Biodiesel production using alumina-supported calcium oxide: An optimization study. Fuel Processing Technology, 91(2), 243-248.

Zong, M.-H., Duan, Z.-Q., Lou, W.-Y., Smith, T.J., Wu, H. 2007. Preparation of a sugar catalyst and its use for highly efficient production of biodiesel. Green Chemistry, 9(5), 434-437.

Table 1. Elemental analyses and surface areas for catalysts 1 – 6.

1. HTC Carbon

%C %H %Oa

%S SA(m2g

-1)

As synthesized 66.54 4.56 28.90 - <10

Sulfonated 53.15 3.02 42.38 1.45 500

3x leached 55.65 3.43 39.85 1.07

2. Pyrolyzed HTC Carbon

As synthesized 97.20 0.32 2.48 - 300

Sulfonated 84.50 0.58 12.67 2.25 ca 1

3. HTC Carbon with PAA

As synthesized 67.11 4.75 28.14 - ca 20

Sulfonated 54.28 2.85 40.94 1.93 ca 1

3x leached 54.07 3.23 41.62 1.08

4. Pyrolyzed HTC Carbon (ZnCl2-insitu + CO2 activation)

As synthesized 94.81 0.34 4.85 - 500

Sulfonated 68.02 1.01 23.78 7.19 ca 2

5. Pyrolyzed HTC Carbon (ZnCl2-impregnated + CO2 activation)

As synthesized 95.76 1.05 3.19 - 1300

Sulfonated 67.52 3.13 28.87 0.48 1160

6. Pyrolyzed Siloxane Resin treated with base

As synthesized 84.90 1.48 a

- 960

Sulfonated 65.37 2.50 a

1.53 880

3x leached 63.55 2.95 a

1.79

aThe oxygen content of the samples is calculated by difference from the other

determinations. This is not possible for the siloxane derived carbons because the residual

silicon content of the samples was not measured.

Table 2: Mercury poisoning experiments carried out on the catalysts leachates as well as

the homogeneous acid catalysts.

Leachate/Homogeneous Acid FFA Conversion (%), 5h

No Hg Hg

Catalyst 1 Leachate 62 13

Catalyst 3 Leachate 45 14

Catalyst 6 Leachate 90 14

Solid Catalyst 1 58 55

Solid Catalyst 3 73 69

Solid Catalyst 6 78 73

0.01 % H2SO4 (v/v) 91 90

1.2 wt.% p-Toluenesulfonic Acid 93 91

Table 3: Initial turnover frequencies of reaction, FFA conversion, and mole percent

catalyst (sulfur) in a reaction charge of the solid catalysts (leached 3x), catalyst leachates

(from the first leaching experiment) and homogeneous acid catalysts. Several solid

carbon catalysts from the literature are shown for comparison.(Fraile et al., 2012; Peng et

al., 2010; Zong et al., 2007)

Solid Acid Catalysts Initial TOF

(s-1

)

% FFA

Conversion at

10 min

Mole %

sulfur

Catalyst 1 0.07 20 0.4

Catalyst 3 0.05 15 0.6

Catalyst 6 0.14 49 0.4

HTC-SO3H (Fraile et al., 2012) 0.017 46(1h) 0.7

“sugar catalyst” (Zong et al.,

2007)

0.005 20(30m) 2

CMK-SO3H (Peng et al., 2010) 0.017 23(1h) 4

Leachates

Catalyst 1 Leachate 33

Catalyst 3 Leachate 24

Catalyst 6 Leachate 39

Homogeneous Catalyst

0.01 % H2SO4 (v/v) 0.20 13 0.1

1.2 wt% p-Toluenesulfonic

Acid

- 84.5 1

Figure Captions:

Figure 1. A: Catalyst 1 Batch Results. Results from the esterification of oleic acid with

catalyst 1. Catalyst contains 1.45 wt% S and has a surface area of 300 m2g

-1. (80°C, 8

hours, 10:1 MeOH:oleic acid, 10% by weight catalyst.) Run 0 represents an unleached

catalyst; Run 1, solid bar, represent conversion after a single extraction; Run 1

crosshatching represents the activity of the leachate. The activity of the leachate obtained

after the third extraction is 10 % conversion which is nearly indistinguishable from

background activity, typically 7-9% conversion in the presence of unsulfonated activated

carbon, represented by the dotted line. B: Catalyst 3 Batch Results. Results from the

esterification of oleic acid with sulfonated 3. Catalyst contains 1.93 wt% S and has a

surface area less than 1 m2g

-1. (80°C, 8 hours, 10:1 MeOH:oleic acid, 10% by weight

catalyst.) C: Catalyst 6 Batch Results. Results from the esterification of oleic acid with

sulfonated 6. Catalyst contains 1.53 wt% S and has a surface area of 960 m2g

-1. (80°C, 8

hours, 10:1 MeOH:oleic acid, 10% by weight catalyst.)

Figure 1.

A. Catalyst 1

B. Catalyst 3

C. Catalyst 6

Sulfonated carbon catalysts prepared from sugar and polyacrylic acid (Cat 3) and siloxane resin (Cat 6)

recycle with little loss in activity after all leachable sites are removed in refluxing methanol.

Cat 3

Cat 60

20

40

60

80

100

12

34

Paper Highlights

Carbon acids are prepared from sulfuric acid and a variety of carbon precursors.

Acid sites are leached to yield stable solid carbon acids with good recyclability.

The leachates contain colloidal carbon acids, which are active for catalysis.

The carbon colloids are observed by TEM.

The carbon colloids are poisoned by metallic mercury.